An assessment of statistical interpolation methods suited for gridded rainfall datasets

Accurate spatial distribution information of rainfall is essential to rainfall-induced hazard predictions and statistical interpolation methods may serve as a useful tool produce detailed from coarse data sources. Although numerous comparison studies about different have been conducted on irregular rain-gauge networks, there need perform such work the increasingly available gridded data. Carried out in Emilia-Romagna region (Italy) 2008 2018, this study aims examine accurate appropriate finer surface maps based 0.25° × ERA5 precipitation datasets. Five techniques, namely Thiessen polygons, Inverse Distance Weighting (IDW), Thin Plate Spline (TPS), Ordinary Kriging (OK) ordinary Co-Kriging (CoK), selected compared at time scales (annual, monthly annual maximum daily precipitation). To assess accuracy, leave-one-out-cross-validation test was used by using indexes Bias, correlation coefficient, Nash-Sutcliffe Efficiency, Root Mean Square Error Kling-Gupta Efficiency (KGE). Additionally, visual inspections are employed evaluate plausibility interpolated maps. Results show: (a) All five certain capabilities improve resolution, but they fail accuracy scale. The OK generally outperforms other four methods, while TPS shows better performance through inspection scale; (b) Unlike case conventional point-based data, multivariate method CoK inferior univariate ones p-power IDW also differs (c) Winter spring results than those summer autumn. This has provided guidance choosing suitable for datasets, which can be expanded regions sources explore generality conclusions.